Objective 3.2 Optimize the Algorithm (e.g., Structure, Run Time, Tuning Hyperparameters) AIP-210 Practice Questions (Page 6)

Questions 26–30

1. 26foundation · easy

   Which of the following is a hyperparameter in a machine learning model?

   • AThe bias term in a linear regression model
   • BThe weights of a neural network after training
   • CThe learning rate used by gradient descent
   • DThe predicted class labels for test samples
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is the core idea behind Bayesian optimization for hyperparameter tuning?

   • AIt uses gradient descent to optimize hyperparameters directly
   • BIt builds a probabilistic model of the objective function to decide where to sample next
   • CIt randomly samples hyperparameters from a uniform distribution
   • DIt exhaustively evaluates all hyperparameter combinations
   Reveal Solution Discussion Select an answer first
4. 28expert · hard

   A team is tuning a model with 10 hyperparameters. They have a fixed budget of 100 training runs. They are considering grid search with 2 values per hyperparameter (2^10 = 1024 combinations) and random search with 100 samples. Which approach is more likely to find a good configuration?

   • AGrid search, because it systematically covers all combinations and is more likely to find the optimum
   • BBoth are equally likely to find a good configuration
   • CRandom search, because it explores a more diverse set of configurations within the budget
   • DGrid search, because it is deterministic and reproducible
   Reveal Solution Discussion Select an answer first
5. 29foundation · easy

   What is a common feature of automated hyperparameter tuning tools like Optuna and Hyperopt?

   • AThey require the user to manually test each hyperparameter combination
   • BThey only work with deep learning models
   • CThey provide built-in search algorithms that automatically explore the hyperparameter space
   • DThey guarantee the optimal hyperparameters are found
   Reveal Solution Discussion Select an answer first
6. 30foundation · easy

   Which of the following is an automated hyperparameter tuning library?

   • AMatplotlib
   • BNumPy
   • CPandas
   • DOptuna
   Reveal Solution Discussion Select an answer first